-# coding: utf-8
-
-from __future__ import unicode_literals
-
import calendar
import copy
import datetime
+import functools
import hashlib
import itertools
import json
import random
import re
import sys
+import threading
import time
import traceback
)
from ..jsinterp import JSInterpreter
from ..utils import (
+ NO_DEFAULT,
+ ExtractorError,
bug_reports_message,
clean_html,
datetime_from_str,
dict_get,
error_to_compat_str,
- ExtractorError,
float_or_none,
format_field,
+ get_first,
int_or_none,
is_html,
join_nonempty,
+ js_to_json,
mimetype2ext,
network_exceptions,
- NO_DEFAULT,
orderedSet,
parse_codecs,
parse_count,
smuggle_url,
str_or_none,
str_to_int,
+ strftime_or_none,
traverse_obj,
try_get,
unescapeHTML,
unified_strdate,
+ unified_timestamp,
unsmuggle_url,
update_url_query,
url_or_none,
variadic,
)
-
-def get_first(obj, keys, **kwargs):
- return traverse_obj(obj, (..., *variadic(keys)), **kwargs, get_all=False)
-
-
# any clients starting with _ cannot be explicity requested by the user
INNERTUBE_CLIENTS = {
'web': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB',
- 'clientVersion': '2.20210622.10.00',
+ 'clientVersion': '2.20211221.00.00',
}
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 1
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB_EMBEDDED_PLAYER',
- 'clientVersion': '1.20210620.0.1',
+ 'clientVersion': '1.20211215.00.01',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 56
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB_REMIX',
- 'clientVersion': '1.20210621.00.00',
+ 'clientVersion': '1.20211213.00.00',
}
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
},
'web_creator': {
- 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
+ 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB_CREATOR',
- 'clientVersion': '1.20210621.00.00',
+ 'clientVersion': '1.20211220.02.00',
}
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
},
'android': {
- 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
+ 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID',
- 'clientVersion': '16.20',
+ 'clientVersion': '16.49',
}
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
'REQUIRE_JS_PLAYER': False
},
'android_embedded': {
- 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
+ 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID_EMBEDDED_PLAYER',
- 'clientVersion': '16.20',
+ 'clientVersion': '16.49',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
'REQUIRE_JS_PLAYER': False
},
'android_music': {
- 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
- 'INNERTUBE_HOST': 'music.youtube.com',
+ 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID_MUSIC',
- 'clientVersion': '4.32',
+ 'clientVersion': '4.57',
}
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
'REQUIRE_JS_PLAYER': False
},
'android_creator': {
+ 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID_CREATOR',
- 'clientVersion': '21.24.100',
+ 'clientVersion': '21.47',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
'REQUIRE_JS_PLAYER': False
},
- # ios has HLS live streams
- # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680
+ # iOS clients have HLS live streams. Setting device model to get 60fps formats.
+ # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
'ios': {
- 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
+ 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'IOS',
- 'clientVersion': '16.20',
+ 'clientVersion': '16.46',
+ 'deviceModel': 'iPhone14,3',
}
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
'REQUIRE_JS_PLAYER': False
},
'ios_embedded': {
- 'INNERTUBE_API_KEY': 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'IOS_MESSAGES_EXTENSION',
- 'clientVersion': '16.20',
+ 'clientVersion': '16.46',
+ 'deviceModel': 'iPhone14,3',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
'REQUIRE_JS_PLAYER': False
},
'ios_music': {
- 'INNERTUBE_API_KEY': 'AIzaSyDK3iBpDP9nHVTk2qL73FLJICfOC3c51Og',
- 'INNERTUBE_HOST': 'music.youtube.com',
+ 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'IOS_MUSIC',
- 'clientVersion': '4.32',
+ 'clientVersion': '4.57',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'IOS_CREATOR',
- 'clientVersion': '21.24.100',
+ 'clientVersion': '21.47',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
# mweb has 'ultralow' formats
# See: https://github.com/yt-dlp/yt-dlp/pull/557
'mweb': {
- 'INNERTUBE_API_KEY': 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8',
+ 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'MWEB',
- 'clientVersion': '2.20210721.07.00',
+ 'clientVersion': '2.20211221.01.00',
}
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 2
},
+ # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
+ # See: https://github.com/zerodytrash/YouTube-Internal-Clients
+ 'tv_embedded': {
+ 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
+ 'INNERTUBE_CONTEXT': {
+ 'client': {
+ 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
+ 'clientVersion': '2.0',
+ },
+ },
+ 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
+ },
}
+def _split_innertube_client(client_name):
+ variant, *base = client_name.rsplit('.', 1)
+ if base:
+ return variant, base[0], variant
+ base, *variant = client_name.split('_', 1)
+ return client_name, base, variant[0] if variant else None
+
+
def build_innertube_clients():
- third_party = {
- 'embedUrl': 'https://google.com', # Can be any valid URL
+ THIRD_PARTY = {
+ 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
}
- base_clients = ('android', 'web', 'ios', 'mweb')
- priority = qualities(base_clients[::-1])
+ BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
+ priority = qualities(BASE_CLIENTS[::-1])
for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
- ytcfg['priority'] = 10 * priority(client.split('_', 1)[0])
-
- if client in base_clients:
- INNERTUBE_CLIENTS[f'{client}_agegate'] = agegate_ytcfg = copy.deepcopy(ytcfg)
- agegate_ytcfg['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
- agegate_ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party
- agegate_ytcfg['priority'] -= 1
- elif client.endswith('_embedded'):
- ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party
+
+ _, base_client, variant = _split_innertube_client(client)
+ ytcfg['priority'] = 10 * priority(base_client)
+
+ if not variant:
+ INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
+ embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
+ embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
+ embedscreen['priority'] -= 3
+ elif variant == 'embedded':
+ ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
ytcfg['priority'] -= 2
else:
ytcfg['priority'] -= 3
_RESERVED_NAMES = (
r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
- r'shorts|movies|results|shared|hashtag|trending|feed|feeds|'
+ r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
r'browse|oembed|get_video_info|iframe_api|s/player|'
r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
_PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
- _NETRC_MACHINE = 'youtube'
+ # _NETRC_MACHINE = 'youtube'
# If True it will raise an error if no login info is provided
_LOGIN_REQUIRED = False
r'(?:www\.)?invidious\.zee\.li',
r'(?:www\.)?invidious\.ethibox\.fr',
r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
+ r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
+ r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
# youtube-dl invidious instances list
r'(?:(?:www|no)\.)?invidiou\.sh',
r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
)
- def _login(self):
- """
- Attempt to log in to YouTube.
- If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
- """
-
- if (self._LOGIN_REQUIRED
- and self.get_param('cookiefile') is None
- and self.get_param('cookiesfrombrowser') is None):
- self.raise_login_required(
- 'Login details are needed to download this content', method='cookies')
- username, password = self._get_login_info()
- if username:
- self.report_warning(f'Cannot login to YouTube using username and password. {self._LOGIN_HINTS["cookies"]}')
-
def _initialize_consent(self):
cookies = self._get_cookies('https://www.youtube.com/')
if cookies.get('__Secure-3PSID'):
consent_id = random.randint(100, 999)
self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
+ def _initialize_pref(self):
+ cookies = self._get_cookies('https://www.youtube.com/')
+ pref_cookie = cookies.get('PREF')
+ pref = {}
+ if pref_cookie:
+ try:
+ pref = dict(compat_urlparse.parse_qsl(pref_cookie.value))
+ except ValueError:
+ self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
+ pref.update({'hl': 'en', 'tz': 'UTC'})
+ self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref))
+
def _real_initialize(self):
+ self._initialize_pref()
self._initialize_consent()
- self._login()
+ self._check_login_required()
+
+ def _check_login_required(self):
+ if (self._LOGIN_REQUIRED
+ and self.get_param('cookiefile') is None
+ and self.get_param('cookiesfrombrowser') is None):
+ self.raise_login_required('Login details are needed to download this content', method='cookies')
_YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
_YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client)
def _extract_context(self, ytcfg=None, default_client='web'):
- _get_context = lambda y: try_get(y, lambda x: x['INNERTUBE_CONTEXT'], dict)
- context = _get_context(ytcfg)
- if context:
- return context
-
- context = _get_context(self._get_default_ytcfg(default_client))
- if not ytcfg:
- return context
-
- # Recreate the client context (required)
- context['client'].update({
- 'clientVersion': self._extract_client_version(ytcfg, default_client),
- 'clientName': self._extract_client_name(ytcfg, default_client),
- })
- visitor_data = try_get(ytcfg, lambda x: x['VISITOR_DATA'], compat_str)
- if visitor_data:
- context['client']['visitorData'] = visitor_data
+ context = get_first(
+ (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
+ # Enforce language and tz for extraction
+ client_context = traverse_obj(context, 'client', expected_type=dict, default={})
+ client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
return context
_SAPISID = None
return None
# SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
sapisidhash = hashlib.sha1(
- f'{time_now} {self._SAPISID} {origin}'.encode('utf-8')).hexdigest()
+ f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
return f'SAPISIDHASH {time_now}_{sapisidhash}'
def _call_api(self, ep, query, video_id, fatal=True, headers=None,
if headers:
real_headers.update(headers)
return self._download_json(
- 'https://%s/youtubei/v1/%s' % (api_hostname or self._get_innertube_host(default_client), ep),
+ f'https://{api_hostname or self._get_innertube_host(default_client)}/youtubei/v1/{ep}',
video_id=video_id, fatal=fatal, note=note, errnote=errnote,
data=json.dumps(data).encode('utf8'), headers=real_headers,
- query={'key': api_key or self._extract_api_key()})
+ query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'})
def extract_yt_initial_data(self, item_id, webpage, fatal=True):
data = self._search_regex(
- (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
+ (fr'{self._YT_INITIAL_DATA_RE}\s*{self._YT_INITIAL_BOUNDARY_RE}',
self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal)
if data:
return self._parse_json(data, item_id, fatal=fatal)
Appears to be used to track session state
"""
return get_first(
- args, (('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))),
+ args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
expected_type=str)
@property
headers['X-Origin'] = origin
return {h: v for h, v in headers.items() if v is not None}
+ def _download_ytcfg(self, client, video_id):
+ url = {
+ 'web': 'https://www.youtube.com',
+ 'web_music': 'https://music.youtube.com',
+ 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
+ }.get(client)
+ if not url:
+ return {}
+ webpage = self._download_webpage(
+ url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
+ return self.extract_ytcfg(video_id, webpage) or {}
+
@staticmethod
def _build_api_continuation_query(continuation, ctp=None):
query = {
warnings.append([alert_type, alert_message])
for alert_type, alert_message in (warnings + errors[:-1]):
- self.report_warning('YouTube said: %s - %s' % (alert_type, alert_message), only_once=only_once)
+ self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
if errors:
raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
if text:
return text
+ def _get_count(self, data, *path_list):
+ count_text = self._get_text(data, *path_list) or ''
+ count = parse_count(count_text)
+ if count is None:
+ count = str_to_int(
+ self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
+ return count
+
+ @staticmethod
+ def _extract_thumbnails(data, *path_list):
+ """
+ Extract thumbnails from thumbnails dict
+ @param path_list: path list to level that contains 'thumbnails' key
+ """
+ thumbnails = []
+ for path in path_list or [()]:
+ for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
+ thumbnail_url = url_or_none(thumbnail.get('url'))
+ if not thumbnail_url:
+ continue
+ # Sometimes youtube gives a wrong thumbnail URL. See:
+ # https://github.com/yt-dlp/yt-dlp/issues/233
+ # https://github.com/ytdl-org/youtube-dl/issues/28023
+ if 'maxresdefault' in thumbnail_url:
+ thumbnail_url = thumbnail_url.split('?')[0]
+ thumbnails.append({
+ 'url': thumbnail_url,
+ 'height': int_or_none(thumbnail.get('height')),
+ 'width': int_or_none(thumbnail.get('width')),
+ })
+ return thumbnails
+
+ @staticmethod
+ def extract_relative_time(relative_time_text):
+ """
+ Extracts a relative time from string and converts to dt object
+ e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
+ """
+ mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text)
+ if mobj:
+ start = mobj.group('start')
+ if start:
+ return datetime_from_str(start)
+ try:
+ return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
+ except ValueError:
+ return None
+
+ def _extract_time_text(self, renderer, *path_list):
+ """@returns (timestamp, time_text)"""
+ text = self._get_text(renderer, *path_list) or ''
+ dt = self.extract_relative_time(text)
+ timestamp = None
+ if isinstance(dt, datetime.datetime):
+ timestamp = calendar.timegm(dt.timetuple())
+
+ if timestamp is None:
+ timestamp = (
+ unified_timestamp(text) or unified_timestamp(
+ self._search_regex(
+ (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
+ text.lower(), 'time text', default=None)))
+
+ if text and timestamp is None:
+ self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True)
+ return timestamp, text
+
def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
default_client='web'):
note='%s%s' % (note, ' (retry #%d)' % count if count else ''))
except ExtractorError as e:
if isinstance(e.cause, network_exceptions):
- if isinstance(e.cause, compat_HTTPError) and not is_html(e.cause.read(512)):
- e.cause.seek(0)
- yt_error = try_get(
- self._parse_json(e.cause.read().decode(), item_id, fatal=False),
- lambda x: x['error']['message'], compat_str)
- if yt_error:
- self._report_alerts([('ERROR', yt_error)], fatal=False)
+ if isinstance(e.cause, compat_HTTPError):
+ first_bytes = e.cause.read(512)
+ if not is_html(first_bytes):
+ yt_error = try_get(
+ self._parse_json(
+ self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
+ lambda x: x['error']['message'], compat_str)
+ if yt_error:
+ self._report_alerts([('ERROR', yt_error)], fatal=False)
# Downloading page may result in intermittent 5xx HTTP error
# Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
# We also want to catch all other network exceptions since errors in later pages can be troublesome
description = self._get_text(renderer, 'descriptionSnippet')
duration = parse_duration(self._get_text(
renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
- view_count_text = self._get_text(renderer, 'viewCountText') or ''
- view_count = str_to_int(self._search_regex(
- r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
- 'view count', default=None))
+ if duration is None:
+ duration = parse_duration(self._search_regex(
+ r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
+ traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
+ video_id, default=None, group='duration'))
+
+ view_count = self._get_count(renderer, 'viewCountText')
uploader = self._get_text(renderer, 'ownerText', 'shortBylineText')
+ channel_id = traverse_obj(
+ renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
+ expected_type=str, get_all=False)
+ timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText')
+ scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
+ overlay_style = traverse_obj(
+ renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
+ get_all=False, expected_type=str)
+ badges = self._extract_badges(renderer)
+ thumbnails = self._extract_thumbnails(renderer, 'thumbnail')
+ navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
+ renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
+ expected_type=str)) or ''
+ url = f'https://www.youtube.com/watch?v={video_id}'
+ if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
+ url = f'https://www.youtube.com/shorts/{video_id}'
return {
'_type': 'url',
'ie_key': YoutubeIE.ie_key(),
'id': video_id,
- 'url': f'https://www.youtube.com/watch?v={video_id}',
+ 'url': url,
'title': title,
'description': description,
'duration': duration,
'view_count': view_count,
'uploader': uploader,
+ 'channel_id': channel_id,
+ 'thumbnails': thumbnails,
+ 'upload_date': (strftime_or_none(timestamp, '%Y%m%d')
+ if self._configuration_arg('approximate_date', ie_key='youtubetab')
+ else None),
+ 'live_status': ('is_upcoming' if scheduled_timestamp is not None
+ else 'was_live' if 'streamed' in time_text.lower()
+ else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges
+ else None),
+ 'release_timestamp': scheduled_timestamp,
+ 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges)
}
youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
(?: # the various things that can precede the ID:
- (?:(?:v|embed|e|shorts)/(?!videoseries)) # v/ or embed/ or e/ or shorts/
+ (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
|(?: # or the v= param in all its forms
(?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
(?:\?|\#!?) # the params delimiter ? or # or #!
'duration': 10,
'view_count': int,
'like_count': int,
- # 'dislike_count': int,
'availability': 'public',
'playable_in_embed': True,
'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
'age_limit': 0,
'start_time': 1,
'end_time': 9,
+ 'channel_follower_count': int
}
},
{
'uploader': 'Philipp Hagemeister',
'uploader_id': 'phihag',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
+ 'channel': 'Philipp Hagemeister',
+ 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
+ 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
'upload_date': '20121002',
- 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
+ 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
'categories': ['Science & Technology'],
'tags': ['youtube-dl'],
'duration': 10,
'view_count': int,
'like_count': int,
- 'dislike_count': int,
+ 'availability': 'public',
+ 'playable_in_embed': True,
+ 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
+ 'live_status': 'not_live',
+ 'age_limit': 0,
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'uploader_id': 'AfrojackVEVO',
'upload_date': '20131011',
'abr': 129.495,
+ 'like_count': int,
+ 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
+ 'playable_in_embed': True,
+ 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
+ 'view_count': int,
+ 'track': 'The Spark',
+ 'live_status': 'not_live',
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
+ 'channel': 'Afrojack',
+ 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
+ 'tags': 'count:19',
+ 'availability': 'public',
+ 'categories': ['Music'],
+ 'age_limit': 0,
+ 'alt_title': 'The Spark',
+ 'channel_follower_count': int
},
'params': {
'youtube_include_dash_manifest': True,
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
'upload_date': '20140605',
'age_limit': 18,
+ 'categories': ['Gaming'],
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
+ 'availability': 'needs_auth',
+ 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
+ 'like_count': int,
+ 'channel': 'The Witcher',
+ 'live_status': 'not_live',
+ 'tags': 'count:17',
+ 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
+ 'playable_in_embed': True,
+ 'view_count': int,
+ 'channel_follower_count': int
},
},
{
'uploader_id': 'FlyingKitty900',
'uploader': 'FlyingKitty',
'age_limit': 18,
+ 'availability': 'needs_auth',
+ 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
+ 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
+ 'channel': 'FlyingKitty',
+ 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
+ 'view_count': int,
+ 'categories': ['Entertainment'],
+ 'live_status': 'not_live',
+ 'tags': ['Flyingkitty', 'godzilla 2'],
+ 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
+ 'like_count': int,
+ 'duration': 177,
+ 'playable_in_embed': True,
+ 'channel_follower_count': int
},
},
{
'id': 'Tq92D6wQ1mg',
'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
'ext': 'mp4',
- 'upload_date': '20191227',
+ 'upload_date': '20191228',
'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
'uploader': 'Projekt Melody',
'description': 'md5:17eccca93a786d51bc67646756894066',
'age_limit': 18,
+ 'like_count': int,
+ 'availability': 'needs_auth',
+ 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
+ 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
+ 'view_count': int,
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
+ 'channel': 'Projekt Melody',
+ 'live_status': 'not_live',
+ 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
+ 'playable_in_embed': True,
+ 'categories': ['Entertainment'],
+ 'duration': 106,
+ 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
+ 'channel_follower_count': int
},
},
{
'uploader_id': 'st3in234',
'description': 'Fan Video. Music & Lyrics by OOMPH!.',
'upload_date': '20130730',
+ 'track': 'Such mich find mich',
+ 'age_limit': 0,
+ 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
+ 'like_count': int,
+ 'playable_in_embed': False,
+ 'creator': 'OOMPH!',
+ 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
+ 'view_count': int,
+ 'alt_title': 'Such mich find mich',
+ 'duration': 210,
+ 'channel': 'Herr Lurik',
+ 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
+ 'categories': ['Music'],
+ 'availability': 'public',
+ 'uploader_url': 'http://www.youtube.com/user/st3in234',
+ 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
+ 'live_status': 'not_live',
+ 'artist': 'OOMPH!',
+ 'channel_follower_count': int
},
},
{
'uploader': 'deadmau5',
'title': 'Deadmau5 - Some Chords (HD)',
'alt_title': 'Some Chords',
+ 'availability': 'public',
+ 'tags': 'count:14',
+ 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
+ 'view_count': int,
+ 'live_status': 'not_live',
+ 'channel': 'deadmau5',
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
+ 'like_count': int,
+ 'track': 'Some Chords',
+ 'artist': 'deadmau5',
+ 'playable_in_embed': True,
+ 'age_limit': 0,
+ 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
+ 'categories': ['Music'],
+ 'album': 'Some Chords',
+ 'channel_follower_count': int
},
'expected_warnings': [
'DASH manifest missing',
'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
'uploader': 'Olympics',
'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
+ 'like_count': int,
+ 'release_timestamp': 1343767800,
+ 'playable_in_embed': True,
+ 'categories': ['Sports'],
+ 'release_date': '20120731',
+ 'channel': 'Olympics',
+ 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
+ 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
+ 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
+ 'age_limit': 0,
+ 'availability': 'public',
+ 'live_status': 'was_live',
+ 'view_count': int,
+ 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
+ 'channel_follower_count': int
},
'params': {
'skip_download': 'requires avconv',
'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
'uploader': '孫ᄋᄅ',
'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
+ 'playable_in_embed': True,
+ 'channel': '孫ᄋᄅ',
+ 'age_limit': 0,
+ 'tags': 'count:11',
+ 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
+ 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
+ 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
+ 'view_count': int,
+ 'categories': ['People & Blogs'],
+ 'like_count': int,
+ 'live_status': 'not_live',
+ 'availability': 'unlisted',
+ 'channel_follower_count': int
},
},
# url_encoded_fmt_stream_map is empty string
'track': 'Dark Walk',
'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
+ 'categories': ['Film & Animation'],
+ 'view_count': int,
+ 'live_status': 'not_live',
+ 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
+ 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
+ 'tags': 'count:13',
+ 'availability': 'public',
+ 'channel': 'IronSoulElf',
+ 'playable_in_embed': True,
+ 'like_count': int,
+ 'age_limit': 0,
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'title': 'md5:e41008789470fc2533a3252216f1c1d1',
'description': 'md5:a677553cf0840649b731a3024aeff4cc',
'duration': 721,
- 'upload_date': '20150127',
+ 'upload_date': '20150128',
'uploader_id': 'BerkmanCenter',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
'uploader': 'The Berkman Klein Center for Internet & Society',
'license': 'Creative Commons Attribution license (reuse allowed)',
+ 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
+ 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
+ 'like_count': int,
+ 'age_limit': 0,
+ 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
+ 'channel': 'The Berkman Klein Center for Internet & Society',
+ 'availability': 'public',
+ 'view_count': int,
+ 'categories': ['Education'],
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
+ 'live_status': 'not_live',
+ 'playable_in_embed': True,
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
'duration': 4060,
- 'upload_date': '20151119',
+ 'upload_date': '20151120',
'uploader': 'Bernie Sanders',
'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
'license': 'Creative Commons Attribution license (reuse allowed)',
+ 'playable_in_embed': True,
+ 'tags': 'count:12',
+ 'like_count': int,
+ 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
+ 'age_limit': 0,
+ 'availability': 'public',
+ 'categories': ['News & Politics'],
+ 'channel': 'Bernie Sanders',
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
+ 'view_count': int,
+ 'live_status': 'not_live',
+ 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'series': 'Mind Field',
'season_number': 1,
'episode_number': 1,
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
+ 'tags': 'count:12',
+ 'view_count': int,
+ 'availability': 'public',
+ 'age_limit': 0,
+ 'channel': 'Vsauce',
+ 'episode': 'Episode 1',
+ 'categories': ['Entertainment'],
+ 'season': 'Season 1',
+ 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
+ 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
+ 'like_count': int,
+ 'playable_in_embed': True,
+ 'live_status': 'not_live',
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'album': 'it\'s too much love to know my dear',
'release_date': '20190313',
'release_year': 2019,
+ 'alt_title': 'Voyeur Girl',
+ 'view_count': int,
+ 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
+ 'playable_in_embed': True,
+ 'like_count': int,
+ 'categories': ['Music'],
+ 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
+ 'channel': 'Stephen',
+ 'availability': 'public',
+ 'creator': 'Stephen',
+ 'duration': 169,
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
+ 'age_limit': 0,
+ 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
+ 'tags': 'count:11',
+ 'live_status': 'not_live',
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'upload_date': '20170613',
'uploader_id': 'ElevageOrVert',
'uploader': 'ElevageOrVert',
+ 'view_count': int,
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
+ 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
+ 'like_count': int,
+ 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
+ 'tags': [],
+ 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
+ 'availability': 'public',
+ 'age_limit': 0,
+ 'categories': ['Pets & Animals'],
+ 'duration': 7,
+ 'playable_in_embed': True,
+ 'live_status': 'not_live',
+ 'channel': 'ElevageOrVert',
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'upload_date': '20130831',
'uploader_id': 'kudvenkat',
'uploader': 'kudvenkat',
+ 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
+ 'like_count': int,
+ 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
+ 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
+ 'live_status': 'not_live',
+ 'categories': ['Education'],
+ 'availability': 'public',
+ 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
+ 'tags': 'count:12',
+ 'playable_in_embed': True,
+ 'age_limit': 0,
+ 'view_count': int,
+ 'duration': 522,
+ 'channel': 'kudvenkat',
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'artist': 'The Cinematic Orchestra',
'track': 'Burn Out',
'album': 'Every Day',
- 'release_data': None,
- 'release_year': None,
+ 'like_count': int,
+ 'live_status': 'not_live',
+ 'alt_title': 'Burn Out',
+ 'duration': 614,
+ 'age_limit': 0,
+ 'view_count': int,
+ 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
+ 'creator': 'The Cinematic Orchestra',
+ 'channel': 'The Cinematic Orchestra',
+ 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
+ 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
+ 'availability': 'public',
+ 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
+ 'categories': ['Music'],
+ 'playable_in_embed': True,
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'ext': 'mp4',
'title': 'San Diego teen commits suicide after bullying over embarrassing video',
'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
- 'uploader': 'CBS This Morning',
+ 'uploader': 'CBS Mornings',
'uploader_id': 'CBSThisMorning',
'upload_date': '20140716',
- 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7'
+ 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
+ 'duration': 170,
+ 'categories': ['News & Politics'],
+ 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
+ 'view_count': int,
+ 'channel': 'CBS Mornings',
+ 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
+ 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
+ 'age_limit': 18,
+ 'availability': 'needs_auth',
+ 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
+ 'like_count': int,
+ 'live_status': 'not_live',
+ 'playable_in_embed': True,
+ 'channel_follower_count': int
}
},
{
'uploader': 'Walk around Japan',
'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
+ 'duration': 1456,
+ 'categories': ['Travel & Events'],
+ 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
+ 'view_count': int,
+ 'channel': 'Walk around Japan',
+ 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
+ 'age_limit': 0,
+ 'availability': 'public',
+ 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
+ 'live_status': 'not_live',
+ 'playable_in_embed': True,
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
'uploader': 'colinfurze',
'uploader_id': 'colinfurze',
'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
- 'description': 'md5:b5096f56af7ccd7a555c84db81738b22'
+ 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
+ 'duration': 596,
+ 'categories': ['Entertainment'],
+ 'uploader_url': 'http://www.youtube.com/user/colinfurze',
+ 'view_count': int,
+ 'channel': 'colinfurze',
+ 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
+ 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
+ 'age_limit': 0,
+ 'availability': 'public',
+ 'like_count': int,
+ 'live_status': 'not_live',
+ 'playable_in_embed': True,
+ 'channel_follower_count': int
},
'params': {
'format': '17', # 3gp format available on android
'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
'upload_date': '20140324',
'uploader': 'SciShow',
+ 'like_count': int,
+ 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
+ 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
+ 'view_count': int,
+ 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
+ 'playable_in_embed': True,
+ 'tags': 'count:12',
+ 'uploader_url': 'http://www.youtube.com/user/scishow',
+ 'availability': 'public',
+ 'channel': 'SciShow',
+ 'live_status': 'not_live',
+ 'duration': 248,
+ 'categories': ['Education'],
+ 'age_limit': 0,
+ 'channel_follower_count': int
}, 'params': {'format': 'mhtml', 'skip_download': True}
- }
+ }, {
+ # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
+ 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
+ 'info_dict': {
+ 'id': '2NUZ8W2llS4',
+ 'ext': 'mp4',
+ 'title': 'The NP that test your phone performance 🙂',
+ 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
+ 'uploader': 'Leon Nguyen',
+ 'uploader_id': 'VNSXIII',
+ 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
+ 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
+ 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
+ 'duration': 21,
+ 'view_count': int,
+ 'age_limit': 0,
+ 'categories': ['Gaming'],
+ 'tags': 'count:23',
+ 'playable_in_embed': True,
+ 'live_status': 'not_live',
+ 'upload_date': '20220103',
+ 'like_count': int,
+ 'availability': 'public',
+ 'channel': 'Leon Nguyen',
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
+ 'channel_follower_count': int
+ }
+ }, {
+ # date text is premiered video, ensure upload date in UTC (published 1641172509)
+ 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
+ 'info_dict': {
+ 'id': 'mzZzzBU6lrM',
+ 'ext': 'mp4',
+ 'title': 'I Met GeorgeNotFound In Real Life...',
+ 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
+ 'uploader': 'Quackity',
+ 'uploader_id': 'QuackityHQ',
+ 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
+ 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
+ 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
+ 'duration': 955,
+ 'view_count': int,
+ 'age_limit': 0,
+ 'categories': ['Entertainment'],
+ 'tags': 'count:26',
+ 'playable_in_embed': True,
+ 'live_status': 'not_live',
+ 'release_timestamp': 1641172509,
+ 'release_date': '20220103',
+ 'upload_date': '20220103',
+ 'like_count': int,
+ 'availability': 'public',
+ 'channel': 'Quackity',
+ 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
+ 'channel_follower_count': int
+ }
+ },
+ { # continuous livestream. Microformat upload date should be preferred.
+ # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
+ 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
+ 'info_dict': {
+ 'id': 'kgx4WGK0oNU',
+ 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
+ 'ext': 'mp4',
+ 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
+ 'availability': 'public',
+ 'age_limit': 0,
+ 'release_timestamp': 1637975704,
+ 'upload_date': '20210619',
+ 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
+ 'live_status': 'is_live',
+ 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
+ 'uploader': '阿鲍Abao',
+ 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
+ 'channel': 'Abao in Tokyo',
+ 'channel_follower_count': int,
+ 'release_date': '20211127',
+ 'tags': 'count:39',
+ 'categories': ['People & Blogs'],
+ 'like_count': int,
+ 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
+ 'view_count': int,
+ 'playable_in_embed': True,
+ 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
+ },
+ 'params': {'skip_download': True}
+ },
]
@classmethod
qs = parse_qs(url)
if qs.get('list', [None])[0]:
return False
- return super(YoutubeIE, cls).suitable(url)
+ return super().suitable(url)
def __init__(self, *args, **kwargs):
- super(YoutubeIE, self).__init__(*args, **kwargs)
+ super().__init__(*args, **kwargs)
self._code_cache = {}
self._player_cache = {}
+ def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data):
+ lock = threading.Lock()
+
+ is_live = True
+ start_time = time.time()
+ formats = [f for f in formats if f.get('is_from_start')]
+
+ def refetch_manifest(format_id, delay):
+ nonlocal formats, start_time, is_live
+ if time.time() <= start_time + delay:
+ return
+
+ _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
+ video_details = traverse_obj(
+ prs, (..., 'videoDetails'), expected_type=dict, default=[])
+ microformats = traverse_obj(
+ prs, (..., 'microformat', 'playerMicroformatRenderer'),
+ expected_type=dict, default=[])
+ _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url)
+ start_time = time.time()
+
+ def mpd_feed(format_id, delay):
+ """
+ @returns (manifest_url, manifest_stream_number, is_live) or None
+ """
+ with lock:
+ refetch_manifest(format_id, delay)
+
+ f = next((f for f in formats if f['format_id'] == format_id), None)
+ if not f:
+ if not is_live:
+ self.to_screen(f'{video_id}: Video is no longer live')
+ else:
+ self.report_warning(
+ f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
+ return None
+ return f['manifest_url'], f['manifest_stream_number'], is_live
+
+ for f in formats:
+ f['is_live'] = True
+ f['protocol'] = 'http_dash_segments_generator'
+ f['fragments'] = functools.partial(
+ self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed)
+
+ def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx):
+ FETCH_SPAN, MAX_DURATION = 5, 432000
+
+ mpd_url, stream_number, is_live = None, None, True
+
+ begin_index = 0
+ download_start_time = ctx.get('start') or time.time()
+
+ lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
+ if lack_early_segments:
+ self.report_warning(bug_reports_message(
+ 'Starting download from the last 120 hours of the live stream since '
+ 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
+ lack_early_segments = True
+
+ known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
+ fragments, fragment_base_url = None, None
+
+ def _extract_sequence_from_mpd(refresh_sequence, immediate):
+ nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
+ # Obtain from MPD's maximum seq value
+ old_mpd_url = mpd_url
+ last_error = ctx.pop('last_error', None)
+ expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403
+ mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
+ or (mpd_url, stream_number, False))
+ if not refresh_sequence:
+ if expire_fast and not is_live:
+ return False, last_seq
+ elif old_mpd_url == mpd_url:
+ return True, last_seq
+ try:
+ fmts, _ = self._extract_mpd_formats_and_subtitles(
+ mpd_url, None, note=False, errnote=False, fatal=False)
+ except ExtractorError:
+ fmts = None
+ if not fmts:
+ no_fragment_score += 2
+ return False, last_seq
+ fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
+ fragments = fmt_info['fragments']
+ fragment_base_url = fmt_info['fragment_base_url']
+ assert fragment_base_url
+
+ _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
+ return True, _last_seq
+
+ while is_live:
+ fetch_time = time.time()
+ if no_fragment_score > 30:
+ return
+ if last_segment_url:
+ # Obtain from "X-Head-Seqnum" header value from each segment
+ try:
+ urlh = self._request_webpage(
+ last_segment_url, None, note=False, errnote=False, fatal=False)
+ except ExtractorError:
+ urlh = None
+ last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
+ if last_seq is None:
+ no_fragment_score += 2
+ last_segment_url = None
+ continue
+ else:
+ should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
+ no_fragment_score += 2
+ if not should_continue:
+ continue
+
+ if known_idx > last_seq:
+ last_segment_url = None
+ continue
+
+ last_seq += 1
+
+ if begin_index < 0 and known_idx < 0:
+ # skip from the start when it's negative value
+ known_idx = last_seq + begin_index
+ if lack_early_segments:
+ known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
+ try:
+ for idx in range(known_idx, last_seq):
+ # do not update sequence here or you'll get skipped some part of it
+ should_continue, _ = _extract_sequence_from_mpd(False, False)
+ if not should_continue:
+ known_idx = idx - 1
+ raise ExtractorError('breaking out of outer loop')
+ last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
+ yield {
+ 'url': last_segment_url,
+ }
+ if known_idx == last_seq:
+ no_fragment_score += 5
+ else:
+ no_fragment_score = 0
+ known_idx = last_seq
+ except ExtractorError:
+ continue
+
+ time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
+
def _extract_player_url(self, *ytcfgs, webpage=None):
player_url = traverse_obj(
ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
get_all=False, expected_type=compat_str)
if not player_url:
return
- if player_url.startswith('//'):
- player_url = 'https:' + player_url
- elif not re.match(r'https?://', player_url):
- player_url = compat_urlparse.urljoin(
- 'https://www.youtube.com', player_url)
- return player_url
+ return urljoin('https://www.youtube.com', player_url)
def _download_player_url(self, video_id, fatal=False):
res = self._download_webpage(
player_id = self._extract_player_info(player_url)
# Read from filesystem cache
- func_id = 'js_%s_%s' % (
- player_id, self._signature_cache_id(example_sig))
+ func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
assert os.path.basename(func_id) == func_id
cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
starts = '' if start == 0 else str(start)
ends = (':%d' % (end + step)) if end + step >= 0 else ':'
steps = '' if step == 1 else (':%d' % step)
- return 's[%s%s%s]' % (starts, ends, steps)
+ return f's[{starts}{ends}{steps}]'
step = None
# Quelch pyflakes warnings - start will be set when step is set
"""Turn the encrypted n field into a working signature"""
if player_url is None:
raise ExtractorError('Cannot decrypt nsig without player_url')
- if player_url.startswith('//'):
- player_url = 'https:' + player_url
- elif not re.match(r'https?://', player_url):
- player_url = compat_urlparse.urljoin(
- 'https://www.youtube.com', player_url)
+ player_url = urljoin('https://www.youtube.com', player_url)
sig_id = ('nsig_value', s)
if sig_id in self._player_cache:
raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id)
def _extract_n_function_name(self, jscode):
- return self._search_regex(
- (r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]{3})\([a-zA-Z0-9]\)',),
- jscode, 'Initial JS player n function name', group='nfunc')
+ nfunc, idx = self._search_regex(
+ r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
+ jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
+ if not idx:
+ return nfunc
+ return json.loads(js_to_json(self._search_regex(
+ rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode,
+ f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)]
def _extract_n_function(self, video_id, player_url):
player_id = self._extract_player_info(player_url)
# cpn generation algorithm is reverse engineered from base.js.
# In fact it works even with dummy cpn.
CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
- cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
+ cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
qs.update({
'ver': ['2'],
def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
return self._parse_json(self._search_regex(
- (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
+ (fr'{regex}\s*{self._YT_INITIAL_BOUNDARY_RE}',
regex), webpage, name, default='{}'), video_id, fatal=False)
- @staticmethod
- def parse_time_text(time_text):
- """
- Parse the comment time text
- time_text is in the format 'X units ago (edited)'
- """
- time_text_split = time_text.split(' ')
- if len(time_text_split) >= 3:
- try:
- return datetime_from_str('now-%s%s' % (time_text_split[0], time_text_split[1]), precision='auto')
- except ValueError:
- return None
-
def _extract_comment(self, comment_renderer, parent=None):
comment_id = comment_renderer.get('commentId')
if not comment_id:
text = self._get_text(comment_renderer, 'contentText')
# note: timestamp is an estimate calculated from the current time and time_text
- time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
- time_text_dt = self.parse_time_text(time_text)
- if isinstance(time_text_dt, datetime.datetime):
- timestamp = calendar.timegm(time_text_dt.timetuple())
+ timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText')
author = self._get_text(comment_renderer, 'authorText')
author_id = try_get(comment_renderer,
lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
_continuation = None
for content in contents:
comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
- expected_comment_count = parse_count(self._get_text(
- comments_header_renderer, 'countText', 'commentsCount', max_runs=1))
+ expected_comment_count = self._get_count(
+ comments_header_renderer, 'countText', 'commentsCount')
if expected_comment_count:
tracker['est_total'] = expected_comment_count
comment_entries_iter = self._comment_entries(
comment_replies_renderer, ytcfg, video_id,
parent=comment.get('id'), tracker=tracker)
- for reply_comment in itertools.islice(comment_entries_iter, min(max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments']))):
- yield reply_comment
+ yield from itertools.islice(comment_entries_iter, min(
+ max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
# Keeps track of counts across recursive calls
if not tracker:
yield from self._comment_entries(renderer, ytcfg, video_id)
max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
- # Force English regardless of account setting to prevent parsing issues
- # See: https://github.com/yt-dlp/yt-dlp/issues/532
- ytcfg = copy.deepcopy(ytcfg)
- traverse_obj(
- ytcfg, ('INNERTUBE_CONTEXT', 'client'), expected_type=dict, default={})['hl'] = 'en'
return itertools.islice(_real_comment_extract(contents), 0, max_comments)
@staticmethod
requested_clients = []
default = ['android', 'web']
allowed_clients = sorted(
- [client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'],
+ (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
for client in self._configuration_arg('player_client'):
if client in allowed_clients:
return orderedSet(requested_clients)
- def _extract_player_ytcfg(self, client, video_id):
- url = {
- 'web_music': 'https://music.youtube.com',
- 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
- }.get(client)
- if not url:
- return {}
- webpage = self._download_webpage(url, video_id, fatal=False, note=f'Downloading {client} config')
- return self.extract_ytcfg(video_id, webpage) or {}
-
def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg):
initial_pr = None
if webpage:
webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
video_id, 'initial player response')
- original_clients = clients
+ all_clients = set(clients)
clients = clients[::-1]
prs = []
- def append_client(client_name):
- if client_name in INNERTUBE_CLIENTS and client_name not in original_clients:
- clients.append(client_name)
+ def append_client(*client_names):
+ """ Append the first client name that exists but not already used """
+ for client_name in client_names:
+ actual_client = _split_innertube_client(client_name)[0]
+ if actual_client in INNERTUBE_CLIENTS:
+ if actual_client not in all_clients:
+ clients.append(client_name)
+ all_clients.add(actual_client)
+ return
# Android player_response does not have microFormats which are needed for
# extraction of some data. So we return the initial_pr with formats
tried_iframe_fallback = False
player_url = None
while clients:
- client = clients.pop()
+ client, base_client, variant = _split_innertube_client(clients.pop())
player_ytcfg = master_ytcfg if client == 'web' else {}
- if 'configs' not in self._configuration_arg('player_skip'):
- player_ytcfg = self._extract_player_ytcfg(client, video_id) or player_ytcfg
+ if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
+ player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
prs.append(pr)
# creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
- if client.endswith('_agegate') and self._is_unplayable(pr) and self.is_authenticated:
- append_client(client.replace('_agegate', '_creator'))
+ if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
+ append_client(f'{base_client}_creator')
elif self._is_agegated(pr):
- append_client(f'{client}_agegate')
+ if variant == 'tv_embedded':
+ append_client(f'{base_client}_embedded')
+ elif not variant:
+ append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
if last_error:
if not len(prs):
self.report_warning(last_error)
return prs, player_url
- def _extract_formats(self, streaming_data, video_id, player_url, is_live):
+ def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration):
itags, stream_ids = {}, []
itag_qualities, res_qualities = {}, {}
q = qualities([
streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
for fmt in streaming_formats:
- if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
+ if fmt.get('targetDurationSec'):
continue
itag = str_or_none(fmt.get('itag'))
itags[itag] = 'https'
stream_ids.append(stream_id)
- tbr = float_or_none(
- fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
+ tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
+ language_preference = (
+ 10 if audio_track.get('audioIsDefault') and 10
+ else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
+ else -1)
+ # Some formats may have much smaller duration than others (possibly damaged during encoding)
+ # Eg: 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
+ # Make sure to avoid false positives with small duration differences.
+ # Eg: __2ABJjxzNo, ySuUZEjARPY
+ is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
+ if is_damaged:
+ self.report_warning(f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
dct = {
'asr': int_or_none(fmt.get('audioSampleRate')),
'filesize': int_or_none(fmt.get('contentLength')),
'format_id': itag,
'format_note': join_nonempty(
'%s%s' % (audio_track.get('displayName') or '',
- ' (default)' if audio_track.get('audioIsDefault') else ''),
+ ' (default)' if language_preference > 0 else ''),
fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
- throttled and 'THROTTLED', delim=', '),
+ throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
'source_preference': -10 if throttled else -1,
'fps': int_or_none(fmt.get('fps')) or None,
'height': height,
'quality': q(quality),
+ 'has_drm': bool(fmt.get('drmFamilies')),
'tbr': tbr,
'url': fmt_url,
'width': int_or_none(fmt.get('width')),
- 'language': audio_track.get('id', '').split('.')[0],
- 'language_preference': 1 if audio_track.get('audioIsDefault') else -1,
+ 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
+ 'desc' if language_preference < -1 else ''),
+ 'language_preference': language_preference,
+ # Strictly de-prioritize damaged and 3gp formats
+ 'preference': -10 if is_damaged else -2 if itag == '17' else None,
}
mime_mobj = re.match(
r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
dct['container'] = dct['ext'] + '_dash'
yield dct
+ live_from_start = is_live and self.get_param('live_from_start')
skip_manifests = self._configuration_arg('skip')
- get_dash = (
- (not is_live or self._configuration_arg('include_live_dash'))
- and 'dash' not in skip_manifests and self.get_param('youtube_include_dash_manifest', True))
- get_hls = 'hls' not in skip_manifests and self.get_param('youtube_include_hls_manifest', True)
+ if not self.get_param('youtube_include_hls_manifest', True):
+ skip_manifests.append('hls')
+ get_dash = 'dash' not in skip_manifests and (
+ not is_live or live_from_start or self._configuration_arg('include_live_dash'))
+ get_hls = not live_from_start and 'hls' not in skip_manifests
def process_manifest_format(f, proto, itag):
if itag in itags:
if process_manifest_format(f, 'dash', f['format_id']):
f['filesize'] = int_or_none(self._search_regex(
r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
+ if live_from_start:
+ f['is_from_start'] = True
+
yield f
def _extract_storyboard(self, player_responses, duration):
spec = get_first(
player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
- if not spec:
+ base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
+ if not base_url:
return
- base_url = spec.pop()
L = len(spec) - 1
for i, args in enumerate(spec):
args = args.split('#')
'width': width,
'height': height,
'fragments': [{
- 'path': url.replace('$M', str(j)),
+ 'url': url.replace('$M', str(j)),
'duration': min(fragment_duration, duration - (j * fragment_duration)),
} for j in range(math.ceil(fragment_count))],
}
- def _real_extract(self, url):
- url, smuggled_data = unsmuggle_url(url, {})
- video_id = self._match_id(url)
-
- base_url = self.http_scheme() + '//www.youtube.com/'
- webpage_url = base_url + 'watch?v=' + video_id
+ def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
webpage = None
if 'webpage' not in self._configuration_arg('player_skip'):
webpage = self._download_webpage(
self._get_requested_clients(url, smuggled_data),
video_id, webpage, master_ytcfg)
+ return webpage, master_ytcfg, player_responses, player_url
+
+ def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
+ live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
+ is_live = get_first(video_details, 'isLive')
+ if is_live is None:
+ is_live = get_first(live_broadcast_details, 'isLiveNow')
+
+ streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
+ formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live, duration))
+
+ return live_broadcast_details, is_live, streaming_data, formats
+
+ def _real_extract(self, url):
+ url, smuggled_data = unsmuggle_url(url, {})
+ video_id = self._match_id(url)
+
+ base_url = self.http_scheme() + '//www.youtube.com/'
+ webpage_url = base_url + 'watch?v=' + video_id
+
+ webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
+
playability_statuses = traverse_obj(
player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
return self.playlist_result(
entries, video_id, video_title, video_description)
- live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
- is_live = get_first(video_details, 'isLive')
- if is_live is None:
- is_live = get_first(live_broadcast_details, 'isLiveNow')
+ duration = int_or_none(
+ get_first(video_details, 'lengthSeconds')
+ or get_first(microformats, 'lengthSeconds')
+ or parse_duration(search_meta('duration'))) or None
- streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
- formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live))
+ live_broadcast_details, is_live, streaming_data, formats = self._list_formats(
+ video_id, microformats, video_details, player_responses, player_url, duration)
if not formats:
if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
if f.get('vcodec') != 'none':
f['stretched_ratio'] = ratio
break
-
- thumbnails = []
- thumbnail_dicts = traverse_obj(
- (video_details, microformats), (..., ..., 'thumbnail', 'thumbnails', ...),
- expected_type=dict, default=[])
- for thumbnail in thumbnail_dicts:
- thumbnail_url = thumbnail.get('url')
- if not thumbnail_url:
- continue
- # Sometimes youtube gives a wrong thumbnail URL. See:
- # https://github.com/yt-dlp/yt-dlp/issues/233
- # https://github.com/ytdl-org/youtube-dl/issues/28023
- if 'maxresdefault' in thumbnail_url:
- thumbnail_url = thumbnail_url.split('?')[0]
- thumbnails.append({
- 'url': thumbnail_url,
- 'height': int_or_none(thumbnail.get('height')),
- 'width': int_or_none(thumbnail.get('width')),
- })
+ thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
thumbnail_url = search_meta(['og:image', 'twitter:image'])
if thumbnail_url:
thumbnails.append({
get_first(video_details, 'channelId')
or get_first(microformats, 'externalChannelId')
or search_meta('channelId'))
- duration = int_or_none(
- get_first(video_details, 'lengthSeconds')
- or get_first(microformats, 'lengthSeconds')
- or parse_duration(search_meta('duration'))) or None
owner_profile_url = get_first(microformats, 'ownerProfileUrl')
live_content = get_first(video_details, 'isLiveContent')
is_live = False
if is_upcoming is None and (live_content or is_live):
is_upcoming = False
- live_starttime = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
- live_endtime = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
- if not duration and live_endtime and live_starttime:
- duration = live_endtime - live_starttime
+ live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
+ live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
+ if not duration and live_end_time and live_start_time:
+ duration = live_end_time - live_start_time
+
+ if is_live and self.get_param('live_from_start'):
+ self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data)
formats.extend(self._extract_storyboard(player_responses, duration))
# URL checking if user don't care about getting the best possible thumbnail
'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
'description': video_description,
- 'upload_date': unified_strdate(
- get_first(microformats, 'uploadDate')
- or search_meta('uploadDate')),
'uploader': get_first(video_details, 'author'),
'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
'uploader_url': owner_profile_url,
'channel_id': channel_id,
- 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
+ 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'),
'duration': duration,
'view_count': int_or_none(
get_first((video_details, microformats), (..., 'viewCount'))
else None if is_live is None or is_upcoming is None
else live_content),
'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL
- 'release_timestamp': live_starttime,
+ 'release_timestamp': live_start_time,
}
pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
})
lang_subs.append({
'ext': fmt,
- 'url': update_url_query(base_url, query),
+ 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
'name': sub_name,
})
subtitles, automatic_captions = {}, {}
for lang_code, caption_track in captions.items():
base_url = caption_track.get('baseUrl')
+ orig_lang = parse_qs(base_url).get('lang', [None])[-1]
if not base_url:
continue
lang_name = self._get_text(caption_track, 'name', max_runs=1)
for trans_code, trans_name in translation_languages.items():
if not trans_code:
continue
+ orig_trans_code = trans_code
if caption_track.get('kind') != 'asr':
+ if 'translated_subs' in self._configuration_arg('skip'):
+ continue
trans_code += f'-{lang_code}'
trans_name += format_field(lang_name, template=' from %s')
- process_language(
- automatic_captions, base_url, trans_code, trans_name, {'tlang': trans_code})
+ # Add an "-orig" label to the original language so that it can be distinguished.
+ # The subs are returned without "-orig" as well for compatibility
+ if lang_code == f'a-{orig_trans_code}':
+ process_language(
+ automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
+ # Setting tlang=lang returns damaged subtitles.
+ process_language(automatic_captions, base_url, trans_code, trans_name,
+ {} if orig_lang == orig_trans_code else {'tlang': trans_code})
info['automatic_captions'] = automatic_captions
info['subtitles'] = subtitles
or self._extract_chapters_from_engagement_panel(initial_data, duration)
or None)
- contents = try_get(
- initial_data,
- lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
- list) or []
- for content in contents:
- vpir = content.get('videoPrimaryInfoRenderer')
- if vpir:
- stl = vpir.get('superTitleLink')
- if stl:
- stl = self._get_text(stl)
- if try_get(
- vpir,
- lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
- info['location'] = stl
- else:
- mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
- if mobj:
- info.update({
- 'series': mobj.group(1),
- 'season_number': int(mobj.group(2)),
- 'episode_number': int(mobj.group(3)),
- })
- for tlb in (try_get(
- vpir,
- lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
- list) or []):
- tbr = tlb.get('toggleButtonRenderer') or {}
- for getter, regex in [(
- lambda x: x['defaultText']['accessibility']['accessibilityData'],
- r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
- lambda x: x['accessibility'],
- lambda x: x['accessibilityData']['accessibilityData'],
- ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
- label = (try_get(tbr, getter, dict) or {}).get('label')
- if label:
- mobj = re.match(regex, label)
- if mobj:
- info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
- break
- sbr_tooltip = try_get(
- vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
- if sbr_tooltip:
- like_count, dislike_count = sbr_tooltip.split(' / ')
+ contents = traverse_obj(
+ initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
+ expected_type=list, default=[])
+
+ vpir = get_first(contents, 'videoPrimaryInfoRenderer')
+ if vpir:
+ stl = vpir.get('superTitleLink')
+ if stl:
+ stl = self._get_text(stl)
+ if try_get(
+ vpir,
+ lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
+ info['location'] = stl
+ else:
+ mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
+ if mobj:
info.update({
- 'like_count': str_to_int(like_count),
- 'dislike_count': str_to_int(dislike_count),
+ 'series': mobj.group(1),
+ 'season_number': int(mobj.group(2)),
+ 'episode_number': int(mobj.group(3)),
})
- vsir = content.get('videoSecondaryInfoRenderer')
- if vsir:
- info['channel'] = self._get_text(vsir, ('owner', 'videoOwnerRenderer', 'title'))
- rows = try_get(
- vsir,
- lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
- list) or []
- multiple_songs = False
- for row in rows:
- if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
- multiple_songs = True
+ for tlb in (try_get(
+ vpir,
+ lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
+ list) or []):
+ tbr = tlb.get('toggleButtonRenderer') or {}
+ for getter, regex in [(
+ lambda x: x['defaultText']['accessibility']['accessibilityData'],
+ r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
+ lambda x: x['accessibility'],
+ lambda x: x['accessibilityData']['accessibilityData'],
+ ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
+ label = (try_get(tbr, getter, dict) or {}).get('label')
+ if label:
+ mobj = re.match(regex, label)
+ if mobj:
+ info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
break
- for row in rows:
- mrr = row.get('metadataRowRenderer') or {}
- mrr_title = mrr.get('title')
- if not mrr_title:
- continue
- mrr_title = self._get_text(mrr, 'title')
- mrr_contents_text = self._get_text(mrr, ('contents', 0))
- if mrr_title == 'License':
- info['license'] = mrr_contents_text
- elif not multiple_songs:
- if mrr_title == 'Album':
- info['album'] = mrr_contents_text
- elif mrr_title == 'Artist':
- info['artist'] = mrr_contents_text
- elif mrr_title == 'Song':
- info['track'] = mrr_contents_text
+ sbr_tooltip = try_get(
+ vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
+ if sbr_tooltip:
+ like_count, dislike_count = sbr_tooltip.split(' / ')
+ info.update({
+ 'like_count': str_to_int(like_count),
+ 'dislike_count': str_to_int(dislike_count),
+ })
+ vsir = get_first(contents, 'videoSecondaryInfoRenderer')
+ if vsir:
+ vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
+ info.update({
+ 'channel': self._get_text(vor, 'title'),
+ 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
+
+ rows = try_get(
+ vsir,
+ lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
+ list) or []
+ multiple_songs = False
+ for row in rows:
+ if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
+ multiple_songs = True
+ break
+ for row in rows:
+ mrr = row.get('metadataRowRenderer') or {}
+ mrr_title = mrr.get('title')
+ if not mrr_title:
+ continue
+ mrr_title = self._get_text(mrr, 'title')
+ mrr_contents_text = self._get_text(mrr, ('contents', 0))
+ if mrr_title == 'License':
+ info['license'] = mrr_contents_text
+ elif not multiple_songs:
+ if mrr_title == 'Album':
+ info['album'] = mrr_contents_text
+ elif mrr_title == 'Artist':
+ info['artist'] = mrr_contents_text
+ elif mrr_title == 'Song':
+ info['track'] = mrr_contents_text
fallbacks = {
'channel': 'uploader',
'channel_id': 'uploader_id',
'channel_url': 'uploader_url',
}
+
+ # The upload date for scheduled, live and past live streams / premieres in microformats
+ # may be different from the stream date. Although not in UTC, we will prefer it in this case.
+ # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
+ upload_date = (
+ unified_strdate(get_first(microformats, 'uploadDate'))
+ or unified_strdate(search_meta('uploadDate')))
+ if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'):
+ upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d')
+ info['upload_date'] = upload_date
+
for to, frm in fallbacks.items():
if not info.get(to):
info[to] = info.get(frm)
class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
+ @staticmethod
+ def passthrough_smuggled_data(func):
+ def _smuggle(entries, smuggled_data):
+ for entry in entries:
+ # TODO: Convert URL to music.youtube instead.
+ # Do we need to passthrough any other smuggled_data?
+ entry['url'] = smuggle_url(entry['url'], smuggled_data)
+ yield entry
+
+ @functools.wraps(func)
+ def wrapper(self, url):
+ url, smuggled_data = unsmuggle_url(url, {})
+ if self.is_music_url(url):
+ smuggled_data['is_music_url'] = True
+ info_dict = func(self, url, smuggled_data)
+ if smuggled_data and info_dict.get('entries'):
+ info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
+ return info_dict
+ return wrapper
+
def _extract_channel_id(self, webpage):
channel_id = self._html_search_meta(
'channelId', webpage, 'channel id', default=None)
def _extract_basic_item_renderer(item):
# Modified from _extract_grid_item_renderer
known_basic_renderers = (
- 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer'
+ 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
)
for key, renderer in item.items():
if not isinstance(renderer, dict):
ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
break
+ def _music_reponsive_list_entry(self, renderer):
+ video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
+ if video_id:
+ return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
+ ie=YoutubeIE.ie_key(), video_id=video_id)
+ playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
+ if playlist_id:
+ video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
+ if video_id:
+ return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
+ ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
+ return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
+ ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
+ browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
+ if browse_id:
+ return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
+ ie=YoutubeTabIE.ie_key(), video_id=browse_id)
+
def _shelf_entries_from_content(self, shelf_renderer):
content = shelf_renderer.get('content')
if not isinstance(content, dict):
# TODO: add support for nested playlists so each shelf is processed
# as separate playlist
# TODO: this includes only first N items
- for entry in self._grid_entries(renderer):
- yield entry
+ yield from self._grid_entries(renderer)
renderer = content.get('horizontalListRenderer')
if renderer:
# TODO
title = self._get_text(shelf_renderer, 'title')
yield self.url_result(shelf_url, video_title=title)
# Shelf may not contain shelf URL, fallback to extraction from content
- for entry in self._shelf_entries_from_content(shelf_renderer):
- yield entry
+ yield from self._shelf_entries_from_content(shelf_renderer)
def _playlist_entries(self, video_list_renderer):
for content in video_list_renderer['contents']:
if video_id:
return self._extract_video(video_renderer)
+ def _hashtag_tile_entry(self, hashtag_tile_renderer):
+ url = urljoin('https://youtube.com', traverse_obj(
+ hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
+ if url:
+ return self.url_result(
+ url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
+
def _post_thread_entries(self, post_thread_renderer):
post_renderer = try_get(
post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
renderer = content.get('backstagePostThreadRenderer')
if not isinstance(renderer, dict):
continue
- for entry in self._post_thread_entries(renderer):
- yield entry
+ yield from self._post_thread_entries(renderer)
r''' # unused
def _rich_grid_entries(self, contents):
if entry:
yield entry
'''
+
def _extract_entries(self, parent_renderer, continuation_list):
# continuation_list is modified in-place with continuation_list = [continuation_token]
continuation_list[:] = [None]
for content in contents:
if not isinstance(content, dict):
continue
- is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
+ is_renderer = traverse_obj(
+ content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
+ expected_type=dict)
if not is_renderer:
renderer = content.get('richItemRenderer')
if renderer:
known_renderers = {
'playlistVideoListRenderer': self._playlist_entries,
'gridRenderer': self._grid_entries,
- 'shelfRenderer': lambda x: self._shelf_entries(x),
+ 'reelShelfRenderer': self._grid_entries,
+ 'shelfRenderer': self._shelf_entries,
+ 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
'backstagePostThreadRenderer': self._post_thread_entries,
'videoRenderer': lambda x: [self._video_entry(x)],
'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
+ 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
}
for key, renderer in isr_content.items():
if key not in known_renderers:
parent_renderer = (
try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
- for entry in extract_entries(parent_renderer):
- yield entry
+ yield from extract_entries(parent_renderer)
continuation = continuation_list[0]
for page_num in itertools.count(1):
headers = self.generate_api_headers(
ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
response = self._extract_response(
- item_id='%s page %s' % (item_id, page_num),
+ item_id=f'{item_id} page {page_num}',
query=continuation, headers=headers, ytcfg=ytcfg,
check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
continue
continuation_renderer = value
continuation_list = [None]
- for entry in known_continuation_renderers[key](continuation_renderer):
- yield entry
+ yield from known_continuation_renderers[key](continuation_renderer)
continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
break
if continuation_renderer:
continue
known_renderers = {
+ 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
'gridPlaylistRenderer': (self._grid_entries, 'items'),
'gridVideoRenderer': (self._grid_entries, 'items'),
'gridChannelRenderer': (self._grid_entries, 'items'),
continue
video_items_renderer = {known_renderers[key][1]: continuation_items}
continuation_list = [None]
- for entry in known_renderers[key][0](video_items_renderer):
- yield entry
+ yield from known_renderers[key][0](video_items_renderer)
continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
break
if video_items_renderer:
break
@staticmethod
- def _extract_selected_tab(tabs):
+ def _extract_selected_tab(tabs, fatal=True):
for tab in tabs:
renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {}
if renderer.get('selected') is True:
return renderer
else:
- raise ExtractorError('Unable to find selected tab')
+ if fatal:
+ raise ExtractorError('Unable to find selected tab')
- @classmethod
- def _extract_uploader(cls, data):
+ def _extract_uploader(self, data):
uploader = {}
- renderer = cls._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
+ renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {}
owner = try_get(
renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
if owner:
- uploader['uploader'] = owner.get('text')
+ owner_text = owner.get('text')
+ uploader['uploader'] = self._search_regex(
+ r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text)
uploader['uploader_id'] = try_get(
owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
uploader['uploader_url'] = urljoin(
def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
playlist_id = title = description = channel_url = channel_name = channel_id = None
- thumbnails_list = []
tags = []
selected_tab = self._extract_selected_tab(tabs)
+ primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
renderer = try_get(
data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
if renderer:
description = renderer.get('description', '')
playlist_id = channel_id
tags = renderer.get('keywords', '').split()
- thumbnails_list = (
- try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
- or try_get(
- self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer'),
- lambda x: x['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
- list)
- or [])
- thumbnails = []
- for t in thumbnails_list:
- if not isinstance(t, dict):
- continue
- thumbnail_url = url_or_none(t.get('url'))
- if not thumbnail_url:
- continue
- thumbnails.append({
- 'url': thumbnail_url,
- 'width': int_or_none(t.get('width')),
- 'height': int_or_none(t.get('height')),
- })
+ # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
+ # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
+ def _get_uncropped(url):
+ return url_or_none((url or '').split('=')[0] + '=s0')
+
+ avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar')
+ if avatar_thumbnails:
+ uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
+ if uncropped_avatar:
+ avatar_thumbnails.append({
+ 'url': uncropped_avatar,
+ 'id': 'avatar_uncropped',
+ 'preference': 1
+ })
+
+ channel_banners = self._extract_thumbnails(
+ data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner']))
+ for banner in channel_banners:
+ banner['preference'] = -10
+
+ if channel_banners:
+ uncropped_banner = _get_uncropped(channel_banners[0]['url'])
+ if uncropped_banner:
+ channel_banners.append({
+ 'url': uncropped_banner,
+ 'id': 'banner_uncropped',
+ 'preference': -5
+ })
+
+ primary_thumbnails = self._extract_thumbnails(
+ primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
+
if playlist_id is None:
playlist_id = item_id
+
+ playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats')
+ last_updated_unix, _ = self._extract_time_text(playlist_stats, 2)
if title is None:
- title = (
- try_get(data, lambda x: x['header']['hashtagHeaderRenderer']['hashtag']['simpleText'])
- or playlist_id)
+ title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) or playlist_id
title += format_field(selected_tab, 'title', ' - %s')
title += format_field(selected_tab, 'expandedText', ' - %s')
+
metadata = {
'playlist_id': playlist_id,
'playlist_title': title,
'uploader': channel_name,
'uploader_id': channel_id,
'uploader_url': channel_url,
- 'thumbnails': thumbnails,
+ 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners,
'tags': tags,
+ 'view_count': self._get_count(playlist_stats, 1),
+ 'availability': self._extract_availability(data),
+ 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'),
+ 'playlist_count': self._get_count(playlist_stats, 0),
+ 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
}
- availability = self._extract_availability(data)
- if availability:
- metadata['availability'] = availability
if not channel_id:
metadata.update(self._extract_uploader(data))
metadata.update({
check_get_keys='contents', fatal=False, ytcfg=ytcfg,
note='Downloading API JSON with unavailable videos')
+ @property
+ def skip_webpage(self):
+ return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
+
def _extract_webpage(self, url, item_id, fatal=True):
retries = self.get_param('extractor_retries', 3)
count = -1
self.report_warning(error_to_compat_str(e))
break
- if dict_get(data, ('contents', 'currentVideoEndpoint')):
+ if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')):
break
last_error = 'Incomplete yt initial data received'
return webpage, data
+ def _report_playlist_authcheck(self, ytcfg, fatal=True):
+ """Use if failed to extract ytcfg (and data) from initial webpage"""
+ if not ytcfg and self.is_authenticated:
+ msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
+ if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
+ raise ExtractorError(
+ f'{msg}. If you are not downloading private content, or '
+ 'your cookies are only for the first account and channel,'
+ ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
+ expected=True)
+ self.report_warning(msg, only_once=True)
+
def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
data = None
- if 'webpage' not in self._configuration_arg('skip'):
+ if not self.skip_webpage:
webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
- if not data:
- if not ytcfg and self.is_authenticated:
- msg = 'Playlists that require authentication may not extract correctly without a successful webpage download.'
- if 'authcheck' not in self._configuration_arg('skip') and fatal:
- raise ExtractorError(
- msg + ' If you are not downloading private content, or your cookies are only for the first account and channel,'
- ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
- expected=True)
+ # Reject webpage data if redirected to home page without explicitly requesting
+ selected_tab = self._extract_selected_tab(traverse_obj(
+ data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {}
+ if (url != 'https://www.youtube.com/feed/recommended'
+ and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
+ and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
+ msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
+ if fatal:
+ raise ExtractorError(msg, expected=True)
self.report_warning(msg, only_once=True)
+ if not data:
+ self._report_playlist_authcheck(ytcfg, fatal=fatal)
data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
return data, ytcfg
return self._extract_response(
item_id=item_id, query=params, ep=ep, headers=headers,
ytcfg=ytcfg, fatal=fatal, default_client=default_client,
- check_get_keys=('contents', 'currentVideoEndpoint'))
+ check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
err_note = 'Failed to resolve url (does the playlist exist?)'
if fatal:
raise ExtractorError(err_note, expected=True)
self.report_warning(err_note, item_id)
- @staticmethod
- def _smuggle_data(entries, data):
- for entry in entries:
- if data:
- entry['url'] = smuggle_url(entry['url'], data)
- yield entry
-
_SEARCH_PARAMS = None
- def _search_results(self, query, params=NO_DEFAULT):
+ def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
data = {'query': query}
if params is NO_DEFAULT:
params = self._SEARCH_PARAMS
if params:
data['params'] = params
+
+ content_keys = (
+ ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
+ ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
+ # ytmusic search
+ ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
+ ('continuationContents', ),
+ )
+ display_id = f'query "{query}"'
+ check_get_keys = tuple({keys[0] for keys in content_keys})
+ ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
+ self._report_playlist_authcheck(ytcfg, fatal=False)
+
continuation_list = [None]
+ search = None
for page_num in itertools.count(1):
data.update(continuation_list[0] or {})
+ headers = self.generate_api_headers(
+ ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
search = self._extract_response(
- item_id='query "%s" page %s' % (query, page_num), ep='search', query=data,
- check_get_keys=('contents', 'onResponseReceivedCommands'))
- slr_contents = try_get(
- search,
- (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
- lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
- list)
- yield from self._extract_entries({'contents': slr_contents}, continuation_list)
+ item_id=f'{display_id} page {page_num}', ep='search', query=data,
+ default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
+ slr_contents = traverse_obj(search, *content_keys)
+ yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
if not continuation_list[0]:
break
'playlist_mincount': 94,
'info_dict': {
'id': 'UCqj7Cz7revf5maW9g5pgNcg',
- 'title': 'Игорь Клейнер - Playlists',
+ 'title': 'Igor Kleiner - Playlists',
'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
- 'uploader': 'Игорь Клейнер',
+ 'uploader': 'Igor Kleiner',
'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
+ 'channel': 'Igor Kleiner',
+ 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
+ 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
+ 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
+ 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
+ 'channel_follower_count': int
},
}, {
'note': 'playlists, multipage, different order',
'playlist_mincount': 94,
'info_dict': {
'id': 'UCqj7Cz7revf5maW9g5pgNcg',
- 'title': 'Игорь Клейнер - Playlists',
+ 'title': 'Igor Kleiner - Playlists',
'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
- 'uploader': 'Игорь Клейнер',
+ 'uploader': 'Igor Kleiner',
+ 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
+ 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
+ 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
+ 'channel': 'Igor Kleiner',
+ 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
+ 'channel_follower_count': int
},
}, {
'note': 'playlists, series',
'description': 'md5:e1384e8a133307dd10edee76e875d62f',
'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
'uploader': '3Blue1Brown',
+ 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
+ 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
+ 'channel': '3Blue1Brown',
+ 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
+ 'tags': ['Mathematics'],
+ 'channel_follower_count': int
},
}, {
'note': 'playlists, singlepage',
'description': 'md5:609399d937ea957b0f53cbffb747a14c',
'uploader': 'ThirstForScience',
'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
+ 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
+ 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
+ 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
+ 'tags': 'count:13',
+ 'channel': 'ThirstForScience',
+ 'channel_follower_count': int
}
}, {
'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
'uploader': 'Sergey M.',
'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
'title': 'youtube-dl public playlist',
+ 'description': '',
+ 'tags': [],
+ 'view_count': int,
+ 'modified_date': '20201130',
+ 'channel': 'Sergey M.',
+ 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
+ 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
+ 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
},
'playlist_count': 1,
}, {
'uploader': 'Sergey M.',
'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
'title': 'youtube-dl empty playlist',
+ 'tags': [],
+ 'channel': 'Sergey M.',
+ 'description': '',
+ 'modified_date': '20160902',
+ 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
+ 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
+ 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
},
'playlist_count': 0,
}, {
'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
'uploader': 'lex will',
'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel': 'lex will',
+ 'tags': ['bible', 'history', 'prophesy'],
+ 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_follower_count': int
},
'playlist_mincount': 2,
}, {
'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
'uploader': 'lex will',
'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'tags': ['bible', 'history', 'prophesy'],
+ 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel': 'lex will',
+ 'channel_follower_count': int
},
'playlist_mincount': 975,
}, {
'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
'uploader': 'lex will',
'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel': 'lex will',
+ 'tags': ['bible', 'history', 'prophesy'],
+ 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_follower_count': int
},
'playlist_mincount': 199,
}, {
'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
'uploader': 'lex will',
'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel': 'lex will',
+ 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'tags': ['bible', 'history', 'prophesy'],
+ 'channel_follower_count': int
},
'playlist_mincount': 17,
}, {
'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
'uploader': 'lex will',
'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel': 'lex will',
+ 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'tags': ['bible', 'history', 'prophesy'],
+ 'channel_follower_count': int
},
'playlist_mincount': 18,
}, {
'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
'uploader': 'lex will',
'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel': 'lex will',
+ 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+ 'tags': ['bible', 'history', 'prophesy'],
+ 'channel_follower_count': int
},
'playlist_mincount': 12,
}, {
'description': 'md5:e1384e8a133307dd10edee76e875d62f',
'uploader': '3Blue1Brown',
'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
+ 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
+ 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
+ 'tags': ['Mathematics'],
+ 'channel': '3Blue1Brown',
+ 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
+ 'channel_follower_count': int
},
}, {
'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
'uploader': 'Christiaan008',
'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
+ 'tags': [],
+ 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
+ 'view_count': int,
+ 'modified_date': '20150605',
+ 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
+ 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
+ 'channel': 'Christiaan008',
},
'playlist_count': 96,
}, {
'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
'uploader': 'Cauchemar',
'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
+ 'channel_url': 'https://www.youtube.com/c/Cauchemar89',
+ 'tags': [],
+ 'modified_date': r're:\d{8}',
+ 'channel': 'Cauchemar',
+ 'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
+ 'view_count': int,
+ 'description': '',
+ 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
},
'playlist_mincount': 1123,
+ 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
}, {
'note': 'even larger playlist, 8832 videos',
'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
'uploader': 'Interstellar Movie',
'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
+ 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
+ 'tags': [],
+ 'view_count': int,
+ 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
+ 'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
+ 'channel': 'Interstellar Movie',
+ 'description': '',
+ 'modified_date': r're:\d{8}',
},
'playlist_mincount': 21,
}, {
'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
'uploader': 'Phim Siêu Nhân Nhật Bản',
'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
+ 'view_count': int,
+ 'channel': 'Phim Siêu Nhân Nhật Bản',
+ 'tags': [],
+ 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
+ 'description': '',
+ 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
+ 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
+ 'modified_date': r're:\d{8}',
},
'playlist_mincount': 200,
+ 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
}, {
'note': 'Playlist with unavailable videos in page 7',
'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
'id': 'UU8l9frL61Yl5KFOl87nIm2w',
'uploader': 'BlankTV',
'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
+ 'channel': 'BlankTV',
+ 'channel_url': 'https://www.youtube.com/c/blanktv',
+ 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
+ 'view_count': int,
+ 'tags': [],
+ 'uploader_url': 'https://www.youtube.com/c/blanktv',
+ 'modified_date': r're:\d{8}',
+ 'description': '',
},
'playlist_mincount': 1000,
+ 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
}, {
'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
'uploader': 'Computerphile',
'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
+ 'uploader_url': 'https://www.youtube.com/user/Computerphile',
+ 'tags': [],
+ 'view_count': int,
+ 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
+ 'channel_url': 'https://www.youtube.com/user/Computerphile',
+ 'channel': 'Computerphile',
},
'playlist_mincount': 11,
}, {
'tags': list,
'view_count': int,
'like_count': int,
- 'dislike_count': int,
},
'params': {
'skip_download': True,
}, {
'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
'info_dict': {
- 'id': '3yImotZU3tw', # This will keep changing
+ 'id': 'GgL890LIznQ', # This will keep changing
'ext': 'mp4',
- 'title': compat_str,
+ 'title': str,
'uploader': 'Sky News',
'uploader_id': 'skynews',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
'upload_date': r're:\d{8}',
- 'description': compat_str,
+ 'description': str,
'categories': ['News & Politics'],
'tags': list,
'like_count': int,
- 'dislike_count': int,
+ 'release_timestamp': 1642502819,
+ 'channel': 'Sky News',
+ 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
+ 'age_limit': 0,
+ 'view_count': int,
+ 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg',
+ 'playable_in_embed': True,
+ 'release_date': '20220118',
+ 'availability': 'public',
+ 'live_status': 'is_live',
+ 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
+ 'channel_follower_count': int
},
'params': {
'skip_download': True,
},
- 'expected_warnings': ['Downloading just video ', 'Ignoring subtitle tracks found in '],
+ 'expected_warnings': ['Ignoring subtitle tracks found in '],
}, {
'url': 'https://www.youtube.com/user/TheYoungTurks/live',
'info_dict': {
'categories': ['News & Politics'],
'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
'like_count': int,
- 'dislike_count': int,
},
'params': {
'skip_download': True,
'info_dict': {
'id': 'cctv9',
'title': '#cctv9',
+ 'tags': [],
},
'playlist_mincount': 350,
}, {
'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
'title': 'NCS Releases',
+ 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
+ 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
+ 'modified_date': r're:\d{8}',
+ 'view_count': int,
+ 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
+ 'tags': [],
+ 'channel': 'NoCopyrightSounds',
},
'playlist_mincount': 166,
+ 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
}, {
'note': 'Topic, should redirect to playlist?list=UU...',
'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
'title': 'Uploads from Royalty Free Music - Topic',
'uploader': 'Royalty Free Music - Topic',
+ 'tags': [],
+ 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
+ 'channel': 'Royalty Free Music - Topic',
+ 'view_count': int,
+ 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
+ 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
+ 'modified_date': r're:\d{8}',
+ 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
+ 'description': '',
},
'expected_warnings': [
- 'A channel/user page was given',
'The URL does not have a videos tab',
+ r'[Uu]navailable videos (are|will be) hidden',
],
'playlist_mincount': 101,
}, {
'info_dict': {
'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
+ 'tags': [],
},
'expected_warnings': [
- 'A channel/user page was given',
- 'The URL does not have a videos tab',
- 'Falling back to channel URL',
+ 'the playlist redirect gave error',
],
'playlist_mincount': 9,
}, {
'info_dict': {
'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
+ 'tags': [],
+ 'view_count': int,
+ 'description': '',
+ 'availability': 'unlisted',
+ 'modified_date': r're:\d{8}',
},
'playlist_count': 50,
}, {
'uploader': 'colethedj',
'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
'title': 'yt-dlp unlisted playlist test',
- 'availability': 'unlisted'
+ 'availability': 'unlisted',
+ 'tags': [],
+ 'modified_date': '20211208',
+ 'channel': 'colethedj',
+ 'view_count': int,
+ 'description': '',
+ 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
+ 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
+ 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
},
'playlist_count': 1,
}, {
'info_dict': {
'id': 'recommended',
'title': 'recommended',
+ 'tags': [],
},
'playlist_mincount': 50,
'params': {
'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
'uploader': 'Cody\'sLab',
'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
+ 'channel': 'Cody\'sLab',
+ 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
+ 'tags': [],
+ 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
+ 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
+ 'channel_follower_count': int
},
'playlist_mincount': 650,
'params': {
'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
'title': 'Uploads from Royalty Free Music - Topic',
'uploader': 'Royalty Free Music - Topic',
+ 'modified_date': r're:\d{8}',
+ 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
+ 'description': '',
+ 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
+ 'tags': [],
+ 'channel': 'Royalty Free Music - Topic',
+ 'view_count': int,
+ 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
},
'expected_warnings': [
- 'A channel/user page was given',
- 'The URL does not have a videos tab',
+ 'does not have a videos tab',
+ r'[Uu]navailable videos (are|will be) hidden',
],
'playlist_mincount': 101,
'params': {
'skip_download': True,
'extractor_args': {'youtubetab': {'skip': ['webpage']}}
},
+ }, {
+ 'note': 'non-standard redirect to regional channel',
+ 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
+ 'only_matching': True
+ }, {
+ 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
+ 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
+ 'info_dict': {
+ 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
+ 'modified_date': '20220407',
+ 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
+ 'tags': [],
+ 'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
+ 'uploader': 'pukkandan',
+ 'availability': 'unlisted',
+ 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
+ 'channel': 'pukkandan',
+ 'description': 'Test for collaborative playlist',
+ 'title': 'yt-dlp test - collaborative playlist',
+ 'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
+ },
+ 'playlist_mincount': 2
}]
@classmethod
def suitable(cls, url):
- return False if YoutubeIE.suitable(url) else super(
- YoutubeTabIE, cls).suitable(url)
+ return False if YoutubeIE.suitable(url) else super().suitable(url)
- def _real_extract(self, url):
- url, smuggled_data = unsmuggle_url(url, {})
- if self.is_music_url(url):
- smuggled_data['is_music_url'] = True
- info_dict = self.__real_extract(url, smuggled_data)
- if info_dict.get('entries'):
- info_dict['entries'] = self._smuggle_data(info_dict['entries'], smuggled_data)
- return info_dict
-
- _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(channel_type)(?P<tab>/\w+))?(?P<post>.*)$')
+ _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
- def __real_extract(self, url, smuggled_data):
+ @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
+ def _real_extract(self, url, smuggled_data):
item_id = self._match_id(url)
url = compat_urlparse.urlunparse(
compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
pre = f'https://www.youtube.com/channel/{item_id}'
+ original_tab_name = tab
if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
# Home URLs should redirect to /videos/
redirect_warning = ('A channel/user page was given. All the channel\'s videos will be downloaded. '
# Handle both video/playlist URLs
qs = parse_qs(url)
- video_id, playlist_id = [qs.get(key, [None])[0] for key in ('v', 'list')]
+ video_id, playlist_id = (qs.get(key, [None])[0] for key in ('v', 'list'))
if not video_id and mobj['not_channel'].startswith('watch'):
if not playlist_id:
data, ytcfg = self._extract_data(url, item_id)
+ # YouTube may provide a non-standard redirect to the regional channel
+ # See: https://github.com/yt-dlp/yt-dlp/issues/2694
+ redirect_url = traverse_obj(
+ data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
+ if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
+ redirect_url = ''.join((
+ urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post']))
+ self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}')
+ return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key())
+
tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list)
if tabs:
selected_tab = self._extract_selected_tab(tabs)
- tab_name = selected_tab.get('title', '')
+ selected_tab_name = selected_tab.get('title', '').lower()
+ if selected_tab_name == 'home':
+ selected_tab_name = 'featured'
+ requested_tab_name = mobj['tab'][1:]
if 'no-youtube-channel-redirect' not in compat_opts:
- if mobj['tab'] == '/live':
+ if requested_tab_name == 'live':
# Live tab should have redirected to the video
raise ExtractorError('The channel is not currently live', expected=True)
- if mobj['tab'] == '/videos' and tab_name.lower() != mobj['tab'][1:]:
- redirect_warning = f'The URL does not have a {mobj["tab"][1:]} tab'
- if not mobj['not_channel'] and item_id[:2] == 'UC':
- # Topic channels don't have /videos. Use the equivalent playlist instead
- pl_id = f'UU{item_id[2:]}'
- pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
- try:
- data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True)
- except ExtractorError:
- redirect_warning += ' and the playlist redirect gave error'
- else:
- item_id, url, tab_name = pl_id, pl_url, mobj['tab'][1:]
- redirect_warning += f'. Redirecting to playlist {pl_id} instead'
- if tab_name.lower() != mobj['tab'][1:]:
- redirect_warning += f'. {tab_name} tab is being downloaded instead'
+ if requested_tab_name not in ('', selected_tab_name):
+ redirect_warning = f'The channel does not have a {requested_tab_name} tab'
+ if not original_tab_name:
+ if item_id[:2] == 'UC':
+ # Topic channels don't have /videos. Use the equivalent playlist instead
+ pl_id = f'UU{item_id[2:]}'
+ pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
+ try:
+ data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
+ except ExtractorError:
+ redirect_warning += ' and the playlist redirect gave error'
+ else:
+ item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name
+ redirect_warning += f'. Redirecting to playlist {pl_id} instead'
+ if selected_tab_name and selected_tab_name != requested_tab_name:
+ redirect_warning += f'. {selected_tab_name} tab is being downloaded instead'
+ else:
+ raise ExtractorError(redirect_warning, expected=True)
if redirect_warning:
- self.report_warning(redirect_warning)
+ self.to_screen(redirect_warning)
self.write_debug(f'Final URL: {url}')
# YouTube sometimes provides a button to reload playlist with unavailable videos.
'info_dict': {
'title': '[OLD]Team Fortress 2 (Class-based LP)',
'id': 'PLBB231211A4F62143',
- 'uploader': 'Wickydoo',
+ 'uploader': 'Wickman',
'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
+ 'view_count': int,
+ 'uploader_url': 'https://www.youtube.com/user/Wickydoo',
+ 'modified_date': r're:\d{8}',
+ 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
+ 'channel': 'Wickman',
+ 'tags': [],
+ 'channel_url': 'https://www.youtube.com/user/Wickydoo',
},
'playlist_mincount': 29,
}, {
'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
'uploader': 'milan',
'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
- }
+ 'description': '',
+ 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
+ 'tags': [],
+ 'modified_date': '20140919',
+ 'view_count': int,
+ 'channel': 'milan',
+ 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
+ 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
+ },
+ 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
}, {
'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
'playlist_mincount': 654,
'uploader': 'LBK',
'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
'description': 'md5:da521864744d60a198e3a88af4db0d9d',
- }
+ 'channel': 'LBK',
+ 'view_count': int,
+ 'channel_url': 'https://www.youtube.com/c/愛低音的國王',
+ 'tags': [],
+ 'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
+ 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
+ 'modified_date': r're:\d{8}',
+ },
+ 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
}, {
'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
'only_matching': True,
qs = parse_qs(url)
if qs.get('v', [None])[0]:
return False
- return super(YoutubePlaylistIE, cls).suitable(url)
+ return super().suitable(url)
def _real_extract(self, url):
playlist_id = self._match_id(url)
'categories': ['Nonprofits & Activism'],
'tags': list,
'like_count': int,
- 'dislike_count': int,
+ 'age_limit': 0,
+ 'playable_in_embed': True,
+ 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
+ 'channel': 'Backus-Page House Museum',
+ 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
+ 'live_status': 'not_live',
+ 'view_count': int,
+ 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
+ 'availability': 'public',
+ 'duration': 59,
},
'params': {
'noplaylist': True,
}), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
+class YoutubeLivestreamEmbedIE(InfoExtractor):
+ IE_DESC = 'YouTube livestream embeds'
+ _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
+ _TESTS = [{
+ 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
+ 'only_matching': True,
+ }]
+
+ def _real_extract(self, url):
+ channel_id = self._match_id(url)
+ return self.url_result(
+ f'https://www.youtube.com/channel/{channel_id}/live',
+ ie=YoutubeTabIE.ie_key(), video_id=channel_id)
+
+
class YoutubeYtUserIE(InfoExtractor):
IE_DESC = 'YouTube user videos; "ytuser:" prefix'
+ IE_NAME = 'youtube:user'
_VALID_URL = r'ytuser:(?P<id>.+)'
_TESTS = [{
'url': 'ytuser:phihag',
ie=YoutubeTabIE.ie_key())
+class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
+ IE_NAME = 'youtube:notif'
+ IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
+ _VALID_URL = r':ytnotif(?:ication)?s?'
+ _LOGIN_REQUIRED = True
+ _TESTS = [{
+ 'url': ':ytnotif',
+ 'only_matching': True,
+ }, {
+ 'url': ':ytnotifications',
+ 'only_matching': True,
+ }]
+
+ def _extract_notification_menu(self, response, continuation_list):
+ notification_list = traverse_obj(
+ response,
+ ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
+ ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
+ expected_type=list) or []
+ continuation_list[0] = None
+ for item in notification_list:
+ entry = self._extract_notification_renderer(item.get('notificationRenderer'))
+ if entry:
+ yield entry
+ continuation = item.get('continuationItemRenderer')
+ if continuation:
+ continuation_list[0] = continuation
+
+ def _extract_notification_renderer(self, notification):
+ video_id = traverse_obj(
+ notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
+ url = f'https://www.youtube.com/watch?v={video_id}'
+ channel_id = None
+ if not video_id:
+ browse_ep = traverse_obj(
+ notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
+ channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
+ post_id = self._search_regex(
+ r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
+ 'post id', default=None)
+ if not channel_id or not post_id:
+ return
+ # The direct /post url redirects to this in the browser
+ url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
+
+ channel = traverse_obj(
+ notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
+ expected_type=str)
+ title = self._search_regex(
+ rf'{re.escape(channel)} [^:]+: (.+)', self._get_text(notification, 'shortMessage'),
+ 'video title', default=None)
+ if title:
+ title = title.replace('\xad', '') # remove soft hyphens
+ upload_date = (strftime_or_none(self._extract_time_text(notification, 'sentTimeText')[0], '%Y%m%d')
+ if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE.ie_key())
+ else None)
+ return {
+ '_type': 'url',
+ 'url': url,
+ 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
+ 'video_id': video_id,
+ 'title': title,
+ 'channel_id': channel_id,
+ 'channel': channel,
+ 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
+ 'upload_date': upload_date,
+ }
+
+ def _notification_menu_entries(self, ytcfg):
+ continuation_list = [None]
+ response = None
+ for page in itertools.count(1):
+ ctoken = traverse_obj(
+ continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
+ response = self._extract_response(
+ item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
+ ep='notification/get_notification_menu', check_get_keys='actions',
+ headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
+ yield from self._extract_notification_menu(response, continuation_list)
+ if not continuation_list[0]:
+ break
+
+ def _real_extract(self, url):
+ display_id = 'notifications'
+ ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
+ self._report_playlist_authcheck(ytcfg)
+ return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
+
+
class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
IE_DESC = 'YouTube search'
IE_NAME = 'youtube:search'
_SEARCH_KEY = 'ytsearch'
_SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
- _TESTS = []
+ _TESTS = [{
+ 'url': 'ytsearch5:youtube-dl test video',
+ 'playlist_count': 5,
+ 'info_dict': {
+ 'id': 'youtube-dl test video',
+ 'title': 'youtube-dl test video',
+ }
+ }]
class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
_SEARCH_KEY = 'ytsearchdate'
IE_DESC = 'YouTube search, newest videos first'
_SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
+ _TESTS = [{
+ 'url': 'ytsearchdate5:youtube-dl test video',
+ 'playlist_count': 5,
+ 'info_dict': {
+ 'id': 'youtube-dl test video',
+ 'title': 'youtube-dl test video',
+ }
+ }]
class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
IE_DESC = 'YouTube search URLs with sorting and filter support'
IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
- _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
+ _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
_TESTS = [{
'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
'playlist_mincount': 5,
'id': 'python',
'title': 'python',
}
-
+ }, {
+ 'url': 'https://www.youtube.com/results?search_query=%23cats',
+ 'playlist_mincount': 1,
+ 'info_dict': {
+ 'id': '#cats',
+ 'title': '#cats',
+ 'entries': [{
+ 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
+ 'title': '#cats',
+ }],
+ },
}, {
'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
'only_matching': True,
return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
-class YoutubeFeedsInfoExtractor(YoutubeTabIE):
+class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
+ IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
+ IE_NAME = 'youtube:music:search_url'
+ _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
+ _TESTS = [{
+ 'url': 'https://music.youtube.com/search?q=royalty+free+music',
+ 'playlist_count': 16,
+ 'info_dict': {
+ 'id': 'royalty free music',
+ 'title': 'royalty free music',
+ }
+ }, {
+ 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
+ 'playlist_mincount': 30,
+ 'info_dict': {
+ 'id': 'royalty free music - songs',
+ 'title': 'royalty free music - songs',
+ },
+ 'params': {'extract_flat': 'in_playlist'}
+ }, {
+ 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
+ 'playlist_mincount': 30,
+ 'info_dict': {
+ 'id': 'royalty free music - community playlists',
+ 'title': 'royalty free music - community playlists',
+ },
+ 'params': {'extract_flat': 'in_playlist'}
+ }]
+
+ _SECTIONS = {
+ 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
+ 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
+ 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
+ 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
+ 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
+ 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
+ }
+
+ def _real_extract(self, url):
+ qs = parse_qs(url)
+ query = (qs.get('search_query') or qs.get('q'))[0]
+ params = qs.get('sp', (None,))[0]
+ if params:
+ section = next((k for k, v in self._SECTIONS.items() if v == params), params)
+ else:
+ section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower()
+ params = self._SECTIONS.get(section)
+ if not params:
+ section = None
+ title = join_nonempty(query, section, delim=' - ')
+ return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
+
+
+class YoutubeFeedsInfoExtractor(InfoExtractor):
"""
Base class for feed extractors
Subclasses must define the _FEED_NAME property.
"""
_LOGIN_REQUIRED = True
- _TESTS = []
+
+ def _real_initialize(self):
+ YoutubeBaseInfoExtractor._check_login_required(self)
@property
def IE_NAME(self):
def _real_extract(self, url):
return self.url_result(
- 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
- ie=YoutubeTabIE.ie_key())
+ f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
class YoutubeWatchLaterIE(InfoExtractor):
def _real_extract(self, url):
video_id = self._match_id(url)
raise ExtractorError(
- 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
+ f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
expected=True)
projects / yt-dlp.git / blobdiff